Counterpoint: Deconstructing PACS Point/Counterpoint

I am neither for or against deconstructed PACS, but rather feel that the way it has been presented is more marketing hype than anything else.

The widespread adoption of standards like Integrating the Healthcare Enterprise (IHE) certainly has helped the concept of deconstructing PACS (DP) to evolve. It seems everywhere you turn there is also talk of interoperability, although today it’s more a buzzword than a reality. More than that, though, the development and growth of alphabet soup submarkets like VNAs, and to a lesser degree medical image sharing (MIS), use of viewer-agnostic zero-footprint viewers (ZFV), and others have helped further the use of DP. That said, there still remain many more turnkey solutions in place than build-it-yourself ones and probably will for quite a while. The reasons for this are many. 

First, the number of vendors who have adopted a DP approach are relatively few, and nearly all of those are smaller independents. None of the majors have adopted a DP philosophy although most majors will connect to other third-party software and solutions if/when pushed.  

Second, few facilities have the internal resources to implement and support third-party solutions from a host of different vendors. VNAs remain the exception because they can support multiple different “ologies,” and both the short- and long-term savings are significant. 

Third, assessing best-of-breed solutions takes time and most facilities don’t have the time to do a detailed assessment and define what best meets their needs. 

And lastly, not all solutions play well with the others. Updating one solution’s software might also have a negative impact on the overall system operation. I have witnessed this time and again when modality manufacturers would update their software without telling the PACS vendors, causing unreadable images the next morning until it got figured out. 

Meshing with an Enterprise Imaging Strategy

Deconstructed PACS is all about sharing across the enterprise, whereby “regular” PACS is more a department-centric solution. This doesn’t mean that one is better than the other, as the solution employed varies based on the needs of the facility. 

VNAs take into effect other “ologies,” while MIS and ZFV both send images across the wide spectrum of clinicians in a facility. Most facilities are looking long-term to the creation of an EHR. While some facilities might elect to just tie the various clinical systems together using an interface engine, others prefer to eliminate as much of the disparate systems using as few applications as possible. This includes using virtual machines (VM) versus physical hardware, the cloud for storage versus on-site archives (or hybrids that use a combination of both), and reference viewers that allow for radiology and cardiology images to both be displayed, as well as reports and so on.

Can regular PACS do that as well as a DP? Yes, but you are limited in both how much you can do and the format therein. Images stored in and retrieved from a VNA offer a lot more flexibility than those stored in a vendor PACS. Images viewed on a ZFP offer more viewing options that those coming from a Web-based PACS. Images transferred to a physician’s office via an MIS offer several advantages over CDs or even USB flash drives. It is important to note that these newer options are not without added cost as many of the “features” you would look at in a DP are included in a regular PACS design. That said, when you look at the bigger picture, oftentimes the benefits outweigh the costs. Once again though, a lot of facilities don’t have the time or resources to make the required evaluation that ensures that everything will work together well and meet the requirements.

The Future of PACS 

There will always be the need for standalone turnkey PACS solutions, especially in the smaller (<150 bed) hospital and diagnostic imaging center markets. Even when DP becomes more widely accepted you’ll still see more turnkey PACS than build-it-yourself PACS in place. 

Simplicity is key and the “one throat to choke” approach usually works best for that. Supporters of DP will say you can choke the throat of a systems integrator as well but it’s just not the same as going back to a PACS vendor saying “It’s your system — make it work,” even if third-party components are in the design as nearly all are. 

The exception to all this will be VNAs. These will become a standard part of PACS in the next five or so years and will no doubt supplant current PACS archives. As VNA prices continue to drop as well as the required data migration costs, VNA adoption will show an even greater uptick. This is predicated on having an accepted industry-wide standard for a VNA adopted, but that should happen fairly soon.  

Lastly, look for data analysis along with the creation of an electronic health record (EHR) to play a strong role in continued PACS growth and utilization. 

PACS has evolved since it was introduced in the 1980s and continues to evolve. As technology advances, this allows new features to be incorporated into PACS system design. While not all advances will be incorporated nor will they happen as quickly as one would like to think, what makes sense from a functional and financial standpoint will no doubt become part of the new PACS.

Editor's note: You can read Tom Watson, Matt Adams and Cris Bennett's Point argument here.

Michael J. Cannavo, better known as the PACSMan, has had over 350 articles on PACS technology published over the past 30 years in industry trade journals. He has extensive experience in the PACS marketplace and has worked with over 300 hospitals, radiology groups and imaging centers nationwide proposing workable, cost-effective solutions for imaging departments, as well as integrating PACS into the EHR.